diff options
| author | jar@chromium.org <jar@chromium.org@0039d316-1c4b-4281-b951-d872f2087c98> | 2009-12-04 22:25:58 +0000 |
|---|---|---|
| committer | jar@chromium.org <jar@chromium.org@0039d316-1c4b-4281-b951-d872f2087c98> | 2009-12-04 22:25:58 +0000 |
| commit | 7446beeb5002262d88844086f0926fc5c9784cfe (patch) | |
| tree | ec075ea9c57b5e7a6f636398909614ad310f0d68 /chrome/app/framework.order | |
| parent | 843c507fd48201618763385037992fada7740861 (diff) | |
| download | chromium_src-7446beeb5002262d88844086f0926fc5c9784cfe.zip chromium_src-7446beeb5002262d88844086f0926fc5c9784cfe.tar.gz chromium_src-7446beeb5002262d88844086f0926fc5c9784cfe.tar.bz2 | |
Delay DNS subresource resolution until navigational resolution is complete
This should probably help some folks that have very sensitive DNS
resolution systems, by at least assuring that the initial navigation
can't get stuck behind congestion induced by DNS prefetching.
There is a tiny chance that this used to be a bit less problematic, as
the DNS prefetch system used to use its own threads, which gave
the navigational resolution a "racing" chance to be handled first (or
at least sooner). With the centralized use of the resolver, the code
(as noted by eroman) actually started the sub-resource resolution
(in all cases) before the navigational resolution. This change
explicitly forces the clearly better ordering.
It won't "fix" bug 3041... but it may help some users with more
recent comments in that bug.
bug=3041
r=eroman
Review URL: http://codereview.chromium.org/464023
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@33869 0039d316-1c4b-4281-b951-d872f2087c98
Diffstat (limited to 'chrome/app/framework.order')
0 files changed, 0 insertions, 0 deletions